Merge tag 'for-upstream' of git://repo.or.cz/qemu/kevin into staging

Block layer patches

- Fix and re-enable GLOBAL_STATE_CODE assertions
- vhost-user: Fixes for VHOST_USER_ADD/REM_MEM_REG
- vmdk: Fix reopening bs->file
- coroutine: use QEMU_DEFINE_STATIC_CO_TLS()
- docs/qemu-img: Fix list of formats which implement check

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmJyi+YRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9aphxAAt0sXEOWlcIeU87NOk+V30rRiBup0K/HZ
# wqsE6e0EMbygmC2aS/xqNu3naQ/TMY6UaoVBWSpf0D3sK2GnWEJW8bjV05ObZBwp
# 6QUgqljk1QAAVv0o2/nViAcV8mEW+OzZLveP+qxFRNlNGoJDsbGzWj939SHM13eu
# ZD+/GGs/qXL3Gxp6adhOBjxbXYjvxm13F3pVjoyAugjMSqoSuCI0eXu1xkwXNHSP
# /wqObH3dQSzIvEXfE/1BOp3ofZwvg+XzeZ6MM4I/lvHDZWuQBfCQcBYKL9mMNWGc
# ijFEeolWt7hER50ik4XPvBmbj0jU2nPXQwo1XcFeWX3MSoNsha2jCZsz4LqzadIN
# YijGQHmkfDRmG2LSoIGcgM7chdwj88K8pfMnrrTsVEB6Dl4QrK6FjXviL5mG+rcX
# 5FbKpgRwm3fmtug7Ttpgm1LJQmwK5A3YPenPH+CC2FoK3Rje46ZoMwQR5PBuHvM1
# rg9RB01eGJQGrw5Rt3VFk7304O/yT2J5m96x6CMejx4CuGK78VpkBC54HixTTh0R
# nXxLLZdqawVqwrPP6sE02FEajM931///nhU6fuN/832m3bYUsfM8SZNVqJh5xoex
# SK8x/a5HnTIkp7kys3f4juc+jzb4Yvka8IBIZi/gqzoqf8POGKLBCTpEXa3lBWIT
# gnCCnWWEdgY=
# =ETW/
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 04 May 2022 09:21:26 AM CDT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]

* tag 'for-upstream' of git://repo.or.cz/qemu/kevin:
  coroutine-win32: use QEMU_DEFINE_STATIC_CO_TLS()
  coroutine: use QEMU_DEFINE_STATIC_CO_TLS()
  coroutine-ucontext: use QEMU_DEFINE_STATIC_CO_TLS()
  iotests/reopen-file: Test reopening file child
  block/vmdk: Fix reopening bs->file
  iotests: Add regression test for issue 945
  Revert "main-loop: Disable GLOBAL_STATE_CODE() assertions"
  qcow2: Do not reopen data_file in invalidate_cache
  block: Classify bdrv_get_flags() as I/O function
  vhost-user: Don't pass file descriptor for VHOST_USER_REM_MEM_REG
  libvhost-user: Fix extra vu_add/rem_mem_reg reply
  docs/vhost-user: Clarifications for VHOST_USER_ADD/REM_MEM_REG
  qemu-img: properly list formats which have consistency check implemented

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

3 files changed, 61 insertions, 36 deletions

diff --git a/util/coroutine-ucontext.c b/util/coroutine-ucontext.c
index ed368e1..ddc98fb 100644
--- a/util/coroutine-ucontext.c
+++ b/util/coroutine-ucontext.c
@@ -25,6 +25,7 @@
 #include "qemu/osdep.h"
 #include <ucontext.h>
 #include "qemu/coroutine_int.h"
+#include "qemu/coroutine-tls.h"
 
 #ifdef CONFIG_VALGRIND_H
 #include <valgrind/valgrind.h>
@@ -66,8 +67,8 @@ typedef struct {
 /**
  * Per-thread coroutine bookkeeping
  */
-static __thread CoroutineUContext leader;
-static __thread Coroutine *current;
+QEMU_DEFINE_STATIC_CO_TLS(Coroutine *, current);
+QEMU_DEFINE_STATIC_CO_TLS(CoroutineUContext, leader);
 
 /*
  * va_args to makecontext() must be type 'int', so passing
@@ -97,14 +98,15 @@ static inline __attribute__((always_inline))
 void finish_switch_fiber(void *fake_stack_save)
 {
 #ifdef CONFIG_ASAN
+    CoroutineUContext *leaderp = get_ptr_leader();
     const void *bottom_old;
     size_t size_old;
 
     __sanitizer_finish_switch_fiber(fake_stack_save, &bottom_old, &size_old);
 
-    if (!leader.stack) {
-        leader.stack = (void *)bottom_old;
-        leader.stack_size = size_old;
+    if (!leaderp->stack) {
+        leaderp->stack = (void *)bottom_old;
+        leaderp->stack_size = size_old;
     }
 #endif
 #ifdef CONFIG_TSAN
@@ -161,8 +163,10 @@ static void coroutine_trampoline(int i0, int i1)
 
     /* Initialize longjmp environment and switch back the caller */
     if (!sigsetjmp(self->env, 0)) {
-        start_switch_fiber_asan(COROUTINE_YIELD, &fake_stack_save, leader.stack,
-                                leader.stack_size);
+        CoroutineUContext *leaderp = get_ptr_leader();
+
+        start_switch_fiber_asan(COROUTINE_YIELD, &fake_stack_save,
+                                leaderp->stack, leaderp->stack_size);
         start_switch_fiber_tsan(&fake_stack_save, self, true); /* true=caller */
         siglongjmp(*(sigjmp_buf *)co->entry_arg, 1);
     }
@@ -297,7 +301,7 @@ qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
     int ret;
     void *fake_stack_save = NULL;
 
-    current = to_;
+    set_current(to_);
 
     ret = sigsetjmp(from->env, 0);
     if (ret == 0) {
@@ -315,18 +319,24 @@ qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
 
 Coroutine *qemu_coroutine_self(void)
 {
-    if (!current) {
-        current = &leader.base;
+    Coroutine *self = get_current();
+    CoroutineUContext *leaderp = get_ptr_leader();
+
+    if (!self) {
+        self = &leaderp->base;
+        set_current(self);
     }
 #ifdef CONFIG_TSAN
-    if (!leader.tsan_co_fiber) {
-        leader.tsan_co_fiber = __tsan_get_current_fiber();
+    if (!leaderp->tsan_co_fiber) {
+        leaderp->tsan_co_fiber = __tsan_get_current_fiber();
     }
 #endif
-    return current;
+    return self;
 }
 
 bool qemu_in_coroutine(void)
 {
-    return current && current->caller;
+    Coroutine *self = get_current();
+
+    return self && self->caller;
 }
diff --git a/util/coroutine-win32.c b/util/coroutine-win32.c
index c196f95..7db2e8f 100644
--- a/util/coroutine-win32.c
+++ b/util/coroutine-win32.c
@@ -24,6 +24,7 @@
 
 #include "qemu/osdep.h"
 #include "qemu/coroutine_int.h"
+#include "qemu/coroutine-tls.h"
 
 typedef struct
 {
@@ -33,8 +34,8 @@ typedef struct
     CoroutineAction action;
 } CoroutineWin32;
 
-static __thread CoroutineWin32 leader;
-static __thread Coroutine *current;
+QEMU_DEFINE_STATIC_CO_TLS(CoroutineWin32, leader);
+QEMU_DEFINE_STATIC_CO_TLS(Coroutine *, current);
 
 /* This function is marked noinline to prevent GCC from inlining it
  * into coroutine_trampoline(). If we allow it to do that then it
@@ -51,7 +52,7 @@ qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
     CoroutineWin32 *from = DO_UPCAST(CoroutineWin32, base, from_);
     CoroutineWin32 *to = DO_UPCAST(CoroutineWin32, base, to_);
 
-    current = to_;
+    set_current(to_);
 
     to->action = action;
     SwitchToFiber(to->fiber);
@@ -88,14 +89,21 @@ void qemu_coroutine_delete(Coroutine *co_)
 
 Coroutine *qemu_coroutine_self(void)
 {
+    Coroutine *current = get_current();
+
     if (!current) {
-        current = &leader.base;
-        leader.fiber = ConvertThreadToFiber(NULL);
+        CoroutineWin32 *leader = get_ptr_leader();
+
+        current = &leader->base;
+        set_current(current);
+        leader->fiber = ConvertThreadToFiber(NULL);
     }
     return current;
 }
 
 bool qemu_in_coroutine(void)
 {
+    Coroutine *current = get_current();
+
     return current && current->caller;
 }
diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
index c03b242..f3e8300 100644
--- a/util/qemu-coroutine.c
+++ b/util/qemu-coroutine.c
@@ -18,6 +18,7 @@
 #include "qemu/atomic.h"
 #include "qemu/coroutine.h"
 #include "qemu/coroutine_int.h"
+#include "qemu/coroutine-tls.h"
 #include "block/aio.h"
 
 /** Initial batch size is 64, and is increased on demand */
@@ -29,17 +30,20 @@ enum {
 static QSLIST_HEAD(, Coroutine) release_pool = QSLIST_HEAD_INITIALIZER(pool);
 static unsigned int pool_batch_size = POOL_INITIAL_BATCH_SIZE;
 static unsigned int release_pool_size;
-static __thread QSLIST_HEAD(, Coroutine) alloc_pool = QSLIST_HEAD_INITIALIZER(pool);
-static __thread unsigned int alloc_pool_size;
-static __thread Notifier coroutine_pool_cleanup_notifier;
+
+typedef QSLIST_HEAD(, Coroutine) CoroutineQSList;
+QEMU_DEFINE_STATIC_CO_TLS(CoroutineQSList, alloc_pool);
+QEMU_DEFINE_STATIC_CO_TLS(unsigned int, alloc_pool_size);
+QEMU_DEFINE_STATIC_CO_TLS(Notifier, coroutine_pool_cleanup_notifier);
 
 static void coroutine_pool_cleanup(Notifier *n, void *value)
 {
     Coroutine *co;
     Coroutine *tmp;
+    CoroutineQSList *alloc_pool = get_ptr_alloc_pool();
 
-    QSLIST_FOREACH_SAFE(co, &alloc_pool, pool_next, tmp) {
-        QSLIST_REMOVE_HEAD(&alloc_pool, pool_next);
+    QSLIST_FOREACH_SAFE(co, alloc_pool, pool_next, tmp) {
+        QSLIST_REMOVE_HEAD(alloc_pool, pool_next);
         qemu_coroutine_delete(co);
     }
 }
@@ -49,27 +53,30 @@ Coroutine *qemu_coroutine_create(CoroutineEntry *entry, void *opaque)
     Coroutine *co = NULL;
 
     if (CONFIG_COROUTINE_POOL) {
-        co = QSLIST_FIRST(&alloc_pool);
+        CoroutineQSList *alloc_pool = get_ptr_alloc_pool();
+
+        co = QSLIST_FIRST(alloc_pool);
         if (!co) {
             if (release_pool_size > qatomic_read(&pool_batch_size)) {
                 /* Slow path; a good place to register the destructor, too.  */
-                if (!coroutine_pool_cleanup_notifier.notify) {
-                    coroutine_pool_cleanup_notifier.notify = coroutine_pool_cleanup;
-                    qemu_thread_atexit_add(&coroutine_pool_cleanup_notifier);
+                Notifier *notifier = get_ptr_coroutine_pool_cleanup_notifier();
+                if (!notifier->notify) {
+                    notifier->notify = coroutine_pool_cleanup;
+                    qemu_thread_atexit_add(notifier);
                 }
 
                 /* This is not exact; there could be a little skew between
                  * release_pool_size and the actual size of release_pool.  But
                  * it is just a heuristic, it does not need to be perfect.
                  */
-                alloc_pool_size = qatomic_xchg(&release_pool_size, 0);
-                QSLIST_MOVE_ATOMIC(&alloc_pool, &release_pool);
-                co = QSLIST_FIRST(&alloc_pool);
+                set_alloc_pool_size(qatomic_xchg(&release_pool_size, 0));
+                QSLIST_MOVE_ATOMIC(alloc_pool, &release_pool);
+                co = QSLIST_FIRST(alloc_pool);
             }
         }
         if (co) {
-            QSLIST_REMOVE_HEAD(&alloc_pool, pool_next);
-            alloc_pool_size--;
+            QSLIST_REMOVE_HEAD(alloc_pool, pool_next);
+            set_alloc_pool_size(get_alloc_pool_size() - 1);
         }
     }
 
@@ -93,9 +100,9 @@ static void coroutine_delete(Coroutine *co)
             qatomic_inc(&release_pool_size);
             return;
         }
-        if (alloc_pool_size < qatomic_read(&pool_batch_size)) {
-            QSLIST_INSERT_HEAD(&alloc_pool, co, pool_next);
-            alloc_pool_size++;
+        if (get_alloc_pool_size() < qatomic_read(&pool_batch_size)) {
+            QSLIST_INSERT_HEAD(get_ptr_alloc_pool(), co, pool_next);
+            set_alloc_pool_size(get_alloc_pool_size() + 1);
             return;
         }
     }